-
Notifications
You must be signed in to change notification settings - Fork 502
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
refined mesh for zero123 setup #164
Comments
Hi! It's definitely possible but I haven't tried yet. Theoretically you just need to copy the |
@bennyguo I'll try that, thanks! |
@bennyguo Hi! I've tried your solution, and changed the parameters of the camera, after that I launched this comand: python launch.py --config=configs/magic3d-refine-sd_custom.yaml --train exp_root_dir="/data/test1/8" here magic3d-refine-sd_custom.yaml is the default magic3d-refine config with the changed camera parameters, but it gives the following error: Do you have any suggestions on how to solve it? |
Hi, I also get the same result as yours. Have you already solved the error? Or if you have any suggestion to write the copy of magic3d-refine-sd.yaml in a correct format? |
@VincentHancoder hi! No! Unfortunately, it’s still unsolved. It seems that you just need to match parameters of the magic3D config with the consideration of the zero123 config. I didn’t have time to do such a deep dive into the code to rewrite the yaml file, there are some mismatches in the parameters that i do not now how to fix. If you, by any chance make it work, please notify me here! |
Sure, I will share my copy format if it can work. But, I don't think it's easy for me as well. If you have any approach, you can share it here, too. |
@VincentHancoder i started with changing camera params as was originally proposed, but found out that there are other mismatches. I tried adding smth from zero123 cfg also, but that solved one issue and produced another. May be if one looks a little deeper into the code, it will be more clear what to change, as I haven’t used this repo extensively. |
Hi!
I would like to know is it possible to apply the same logic as shown in readme in the Magic3D section, but for training with zero123?
In the given examples there is a text prompt, but is it possible to use an image? for example, can i first do: python launch.py --config configs/zero123.yaml and then after it's done somehow refine the trained nerf? And if it's possible, what would be the recomended way to do this?
thanks in advance!
The text was updated successfully, but these errors were encountered: